I know each environment is different, but there is a "loop" between workarounds, hence I disagree with Cisco's approach to this issue and instead should focus on addressing the bug.
Example: CSCvr17987's "Workaround: Opt not to cluster the Expressway nodes". This is not a true solution since you will run into other bugs, such as https://bst.cloudapps.cisco.com/bugsearch/bug/CSCvj49486 - "IP Phones over MRA do not support failover without clustered Expressway C and E", which has "Workaround: Cluster Expressway C & E to allow for failover".